Ionic Liquids as Entrainers in Extractive Distillation - American

Nov 9, 2006 - will come down in price when its demand augments, and what is more, one should not forget that the ionic liquid consumption in the disti...
0 downloads 0 Views 76KB Size
J. Chem. Eng. Data 2007, 52, 141-147

141

Ionic Liquids as Entrainers in Extractive Distillation: Isobaric Vapor-Liquid Equilibria for Acetone + Methanol + 1-Ethyl-3-methylimidazolium Trifluoromethanesulfonate A. Vicent Orchille´ s, Pablo J. Miguel, Ernesto Vercher, and Antoni Martı´nez-Andreu* Departamento de Ingenierı´a Quı´mica, Escuela Te´cnica Superior de Ingenierı´a, Universitat de Vale`ncia, 46100 Burjassot, Valencia, Spain

Isobaric vapor-liquid equilibria for the binary systems acetone + methanol, acetone + 1-ethyl-3-methylimidazolium trifluoromethanesulfonate ([emim][triflate]), and methanol + [emim][triflate] as well as the vapor-liquid equilibrium for the acetone + methanol + [emim][triflate] ternary system have been obtained at 100 kPa using a recirculating still. The addition of the ionic liquid to the solvent mixture produced an important salting-out effect, and the azeotrope tended to disappear for small contents of ionic liquid. The experimental binary data sets were independently fitted with the electrolyte NRTL model, and the parameters of Mock’s model were estimated for each binary system. These parameters were used to predict the ternary vapor-liquid equilibrium which agreed very well with the experimental one.

Introduction Generally, the term “ionic liquid” (IL) stands for a class of chemicals composed entirely of ions with very low melting points (mainly below 100 °C). Typically, ionic liquids comprise an organic greatly asymmetric substituted cation (imidazolium, pyridinium, pyrrolidinium, quaternary ammonium, tetraalkylphosphonium, etc.) and an inorganic weak anion (halide, nitrate, acetate, trifluoromethanesulfonate, tetrafluoroborate, hexafluorophosphate, etc.). These cations, substituents, and anions can be virtually varied at will to change their chemical and physical properties.1 Because of their structure and ionic interactions, they present unique properties:2 they have a liquidus range of 300 °C, are outstandingly good solvents for a wide range of inorganic, organic, and polymeric materials, exhibit Bronsted, Lewis, and Franklin acidity as well as superacidity, have no effective vapor pressure, and are thermally stable up to 200 °C. Although ILs have been known for 100 years, it is only during the past decade that the interest in ILs has increased dramatically. Applications as “green” solvent replacing volatile organic compounds, catalyst in chemical reactions, extraction media for separation processes, lubricants, thermofluids, plasticizers, and electrically conductive liquids in electrochemistry have been reported.3 Ionic liquids, when dissolved in solvent mixtures, like inorganic salts, are dissociated in ions which interact with the components of the liquid mixture modifying its chemical potential. In cases where an IL interacts more strongly with one solvent in a mixture, azeotropes could be broken and improved separations might be achieved. Even when this effect is not too intense, the greater solubility of ILs in low-polar solvents allows a greater electrolyte concentration in the mixture and therefore a stronger salt effect. From the viewpoint of separation processes, when an ionic liquid is used as an entrainer in extractive distillation of hardly * To whom correspondence should be addressed. Fax: +34 963 544 898. E-mail: [email protected].

separating solvent mixtures, it is withdrawn as the column bottom’s effluent with the heavier components, from which the ionic liquid can be easily and completely separated by flash distillation because it presents a null vapor pressure. Moreover, unlike inorganic salts, ILs are liquids at work temperatures and can be easily recirculated and mixed with the reflux stream introducing none of the less-volatile components at the top of the distillation column. Because of that, the choice of an ionic liquid azeotrope breaker could have significant processing advantages over the use of solid inorganic salts or liquid entrainers to shift or break azeotropes.4 Arlt and co-workers5-7 were the first group that suggested ionic liquids as selective compounds for separation of azeotropic mixtures.8,9 They reported the effect of various ionic liquids on ethanol + water,8-10 acetone + methanol,8 water + acetic acid,8 tetrahydrofurane + water,9-11 methanol + trimethylborate,12 and 1-hexene + n-hexane13 azeotropic systems. Nevertheless, even though the use of ionic liquids as solvents or entrainers in separation technology is promising, thermodynamic data are rare and currently only a few investigations on vapor-liquid equilibria have been accomplished. Most works on the vapor-liquid equilibrium of IL-containing systems are limited to determining the vapor pressure and/or activity coefficients of one or two solvents in ILs. Among them, we should cite here those presented by Anthony et al.,14 Heintz and co-workers,15-20 Lee and co-workers,21,22 Gmehling and coworkers,23-26 Abusen et al.,27 Orchille´s et al.,28 and Banerjee et al.29 Likewise, the results of Arlt and co-workers quoted before8-13 were obtained using headspace chromatography techniques at constant temperature. Only Zhao et al.30 have reported complete isobaric vapor-liquid equilibria (T-x′-y) for the ethanol-water system containing different ILs. Our research group has been working for 15 years on the effect of different inorganic salts on the vapor-liquid equilibrium of hardly separating solvent mixtures, mainly alcohols + water. Currently, we have started a new research line using ionic liquids instead of inorganic salts so as to modify the vaporliquid equilibrium of solvent mixtures resulting in difficult

10.1021/je0603170 CCC: $37.00 © 2007 American Chemical Society Published on Web 11/09/2006

142

Journal of Chemical and Engineering Data, Vol. 52, No. 1, 2007

Table 1. Vapor-Liquid Equilibrium Data for Acetone (1) + Methanol (2) at 100 kPa

Table 2. Vapor-Liquid Equilibrium Data for Acetone (1) + [emim][triflate] (3) at 100 kPa

x1

y1

T/K

x3

T/K

0.000 0.015 0.029 0.057 0.097 0.137 0.176 0.214 0.254 0.292 0.331 0.372 0.412 0.449 0.491 0.531 0.574 0.617 0.659 0.697 0.727 0.743 0.760 0.781 0.800 0.820 0.841 0.862 0.883 0.930 0.972 1.000

0.000 0.034 0.063 0.119 0.188 0.247 0.303 0.348 0.388 0.423 0.456 0.489 0.518 0.546 0.576 0.603 0.630 0.662 0.690 0.718 0.739 0.751 0.765 0.781 0.797 0.813 0.830 0.850 0.870 0.918 0.964 1.000

337.41 336.90 336.45 335.60 334.52 333.58 332.72 332.05 331.45 330.95 330.48 330.06 329.70 329.39 329.10 328.86 328.66 328.43 328.31 328.20 328.15 328.13 328.13 328.13 328.14 328.15 328.18 328.23 328.29 328.49 328.75 329.01

0.0157 0.0263 0.0395 0.0544 0.0700 0.0832 0.1004 0.1150 0.1282 0.1448 0.1585 0.1721 0.1867 0.2010 0.2124 0.2240 0.2359 0.2481 0.2596 0.2702 0.2769 0.2866 0.2945 0.3095 0.3236 0.3435 0.3532 0.3636 0.3686 0.3791 0.3908 0.4022 0.4105 0.4125 0.4282 0.4349 0.4399 0.4513 0.4609 0.4722 0.4876

329.45 329.69 330.01 330.34 330.69 331.01 331.38 331.73 332.05 332.55 332.85 333.25 333.75 334.15 334.65 334.95 335.35 335.85 336.30 336.65 337.05 337.45 337.95 338.45 339.15 340.25 340.60 341.21 341.45 342.05 342.77 343.45 343.55 344.05 344.95 345.45 345.95 346.45 347.25 347.95 348.95

separation by distillation. In this work, complete isobaric vapor-liquid equilibria data for acetone + methanol + 1-ethyl-3-methylimidazolium trifluoromethanesulfonate ([emim][triflate]) binary and ternary systems are presented at 100 kPa. In a previous work,31 we studied the effect of an inorganic salt (lithium nitrate) on the vapor-liquid equilibrium of the acetone + methanol mixture.

Experimental Section Materials. The chemicals used were acetone (Merck, GR grade, minimum 99.5 wt %), dried methanol (Riedel-de Hae¨n, Analytical Reagent, minimum 99.8 wt %), and the ionic liquid 1-ethyl-3-methylimidazolium trifluoromethanesulfonate supplied by Solvent Innovation (Purum, minimum 98 wt %). Because of its hygroscopic character, the ionic liquid was desiccated at 10-3 Torr of vacuum overnight. No impurities were detected by GC. The solvents were directly used without further purification. Apparatus and Procedure. The equilibrium apparatus used to obtain the experimental data was an all-glass dynamic recirculating still manufactured by Fischer (Labodest model). In this apparatus, the vaporization ensues via an electrical immersion heater. Because of the regular circulation of both phases and the simultaneous mixing of the back-flowing circulation streams in a special mixing chamber (Cottrell pump),32 a quick adjustment of the equilibrium will be obtained. Before entering the separation chamber, the vapor-liquid mixture passes an extended contact line, which guarantees an intense phase exchange. The construction of the separation chamber prevents an entrainment of liquid particles into the vapor phase. The still is capable of handling pressures from (0.25 to 400) kPa and temperatures up to 523 K. The apparatus pressure was kept constant by means of a vacuum pump and an electrovalve

modified by an on-off pressure controller. The equilibrium pressure was measured with a Fischer vacuum probe, whereas the boiling point was measured with a Fluke 1502A digital thermometer and a Pt-100 probe. The standard uncertainties for pressure and temperature measurements were 0.01 kPa and 0.01 K, respectively. Every experimental point of the binary acetone + methanol system was obtained from an initial sample of pure acetone at which different quantities of methanol were added, whereas for the acetone + IL and methanol + IL binary systems, acetone or methanol, respectively, was added to a concentrated solution until a very diluted solution was achieved. For the ternary system, several acetone + IL mixtures of distinct composition were taken and different quantities of a mixture of methanol + IL having a slightly higher concentration than the original one were added. The samples were prepared with a Mettler AE200 analytical balance with a standard uncertainty of 0.0001 g. Only when constant temperature was reached (30 min or longer) were the equilibrium conditions assumed. Sample Analysis. Compositions of the condensed vapor (acetone + methanol) and liquid (acetone + methanol + IL) phases were analyzed by using a Varian Star 3400 CX gas chromatograph with a thermal conductivity detector (TCD). The chromatographic column (2 m × 0.125 in.) was packed with Porapak QS. The carrier gas was helium flowing at 30 cm3‚min-1, and the operating conditions were as follows: injector and oven temperatures of 453 K and a detector

Journal of Chemical and Engineering Data, Vol. 52, No. 1, 2007 143 Table 3. Vapor-Liquid Equilibrium Data for Methanol (2) + [emim][triflate] (3) at 100 kPa

Table 4. Vapor-Liquid Equilibrium Data for Acetone (1) + Methanol (2) + [emim][triflate] (3) at 100 kPa

x3

T/K

x3

x′1

y1

T/K

x3

x′1

y1

T/K

0.0030 0.0061 0.0117 0.0192 0.0281 0.0373 0.0480 0.0563 0.0649 0.0774 0.0903 0.1048 0.1126 0.1247 0.1352 0.1447 0.1523 0.1612 0.1669 0.1798 0.1866 0.1980 0.2110 0.2228 0.2334 0.2455 0.2564 0.2687 0.2782 0.2921 0.3065 0.3246 0.3442 0.3656 0.3912 0.4235 0.4504 0.4743

337.58 337.68 337.87 338.10 338.36 338.63 338.94 339.19 339.42 339.81 340.21 340.68 340.90 341.37 341.69 342.05 342.30 342.70 342.88 343.30 343.69 344.13 344.67 345.19 345.67 346.21 346.70 347.32 347.78 348.45 349.35 350.35 351.55 352.75 354.55 356.55 358.35 360.15

0.056 0.057 0.058 0.059 0.059 0.060 0.060 0.060 0.060 0.059 0.059 0.059 0.058 0.057 0.057 0.056 0.055 0.054 0.125 0.123 0.121 0.123 0.123 0.123 0.124 0.125 0.122 0.121 0.120 0.120 0.119 0.118 0.119 0.117 0.116 0.117 0.114 0.116 0.115 0.167 0.163 0.167 0.169 0.168 0.168 0.171 0.172 0.166

0.000 0.034 0.074 0.120 0.174 0.250 0.316 0.388 0.455 0.530 0.596 0.675 0.749 0.815 0.875 0.932 0.980 1.000 0.000 0.016 0.082 0.122 0.163 0.209 0.259 0.306 0.358 0.399 0.464 0.520 0.578 0.632 0.689 0.758 0.817 0.873 0.924 0.979 1.000 0.000 0.021 0.057 0.085 0.123 0.160 0.201 0.248 0.298

0.000 0.069 0.137 0.206 0.276 0.367 0.430 0.492 0.550 0.613 0.662 0.723 0.780 0.833 0.884 0.934 0.981 1.000 0.000 0.035 0.147 0.204 0.263 0.326 0.380 0.428 0.479 0.522 0.570 0.619 0.668 0.710 0.757 0.803 0.850 0.895 0.936 0.983 1.000 0.000 0.046 0.104 0.147 0.201 0.250 0.302 0.356 0.412

339.19 338.09 337.11 336.09 335.03 333.75 332.83 332.02 331.40 330.87 330.49 330.17 329.97 329.89 329.90 330.01 330.18 330.30 341.37 340.60 338.90 338.13 337.39 336.60 335.80 335.18 334.47 333.93 333.37 332.89 332.48 332.18 331.94 331.73 331.64 331.62 331.56 331.66 331.73 342.88 342.01 341.34 340.78 340.05 339.30 338.75 338.11 336.96

0.168 0.168 0.167 0.166 0.163 0.162 0.163 0.158 0.165 0.160 0.169 0.168 0.160 0.256 0.247 0.254 0.259 0.262 0.258 0.259 0.254 0.256 0.258 0.258 0.255 0.253 0.250 0.249 0.248 0.247 0.248 0.355 0.360 0.365 0.368 0.370 0.367 0.367 0.367 0.365 0.363 0.362 0.361 0.356 0.355 0.351 0.358

0.345 0.398 0.459 0.522 0.581 0.636 0.696 0.746 0.829 0.889 0.927 0.979 1.000 0.000 0.032 0.116 0.163 0.212 0.314 0.364 0.418 0.466 0.516 0.585 0.689 0.738 0.790 0.839 0.892 0.940 1.000 0.000 0.048 0.092 0.138 0.189 0.258 0.323 0.395 0.469 0.536 0.610 0.686 0.756 0.831 0.900 1.000

0.457 0.513 0.564 0.623 0.670 0.718 0.764 0.804 0.865 0.911 0.943 0.984 1.000 0.000 0.059 0.187 0.250 0.308 0.423 0.472 0.522 0.572 0.620 0.663 0.751 0.801 0.841 0.880 0.921 0.956 1.000 0.000 0.083 0.151 0.212 0.277 0.357 0.431 0.496 0.570 0.642 0.703 0.770 0.825 0.886 0.937 1.000

336.43 335.83 335.29 334.65 334.14 333.72 333.35 333.11 333.25 332.93 333.13 333.05 332.93 346.70 345.45 343.97 343.40 342.80 340.92 340.29 339.77 339.33 338.70 338.25 337.30 336.86 336.66 336.30 336.02 335.78 335.85 352.05 351.20 350.66 350.04 349.33 348.13 347.14 346.22 345.35 344.52 343.71 342.72 342.02 341.46 340.85 340.62

temperature of 493 K. A calibration curve was obtained from a set of gravimetrically prepared standard solutions which allowed us to quantify the amounts of methanol and acetone in the samples. Thus, the combined standard uncertainty of the measured mole fraction in the vapor phase was 0.001. The IL mole fraction content in the liquid phase was gravimetrically determined after the volatile components were separated from a known mass of sample by evaporation at 393 K until constant weight. In this way, the combined standard uncertainty of the mole fraction of acetone, methanol, and [emim][triflate] was 0.001.

Table 5. Estimated Values of Nonrandomness Factors, ri,j, and Energy Parameters, ∆gi,j and ∆gj,i, for the Electrolyte NRTL Model i component

Results and Discussion Experimental Data. Vapor-liquid equilibria for the binary systems acetone (1) + methanol (2), acetone (1) + [emim][triflate] (3), and methanol (2) + [emim][triflate] (3) as well as the vapor-liquid equilibrium for the acetone (1) + methanol (2) + [emim][triflate] (3) ternary system have been obtained at 100 kPa, and the results are shown in Tables 1 to 4. In these tables, xi represents the component i mole fraction in the liquid phase and x′1 represents the mole fraction of acetone in the liquid phase expressed on an IL-free basis. y1 represents the mole fraction of acetone in the vapor phase, and T is the equilibrium temperature. The acetone (1) + methanol (2) system shows a minimum point azeotrope at x1 ) 0.781 and T ) 328.13 K, which can be interpolated from experimental data. These results are in accordance with those reported by Harper and Moore33 and Freshwater and Pike34 at the same pressure. Besides this, the

acetone acetone methanol a

∆gi,j

∆gj,i

J‚mol-1

J‚mol-1

j component

Ri,j

methanol [emim][triflate] [emim][triflate]

0.300a

924.2a

0.335b 0.389b

12765.9b 11950.6b

From Vercher et al.31

b

863.1a -4468.1b -4547.0b

From this work.

system shows a good thermodynamic consistency according to the Van Ness test35 modified by Fredenslund.36 No consistency test was used for IL-containing systems. Calculation of the Phase Equilibrium. Because of the fact that ionic liquids are organic salts and exhibit ionic character, electrolyte models have been used to correlate the vapor-liquid equilibrium of solvent + IL systems. So, Arlt and coworkers,8,10,12 Heintz and co-workers,15-17,19,20 and Gmehling and co-workers23,24 have used the electrolyte NRTL model, and Gmehling and co-workers have also used the electrolyte UNIQUAC model23-26 and the electrolyte Wilson model.24 In

144

Journal of Chemical and Engineering Data, Vol. 52, No. 1, 2007

Table 6. Mean Absolute Deviations δy and δT and Standard Deviations σy and σT between Experimental and Calculated Values of the Vapor-Phase Mole Fractions and the Equilibrium Temperatures

c

δTc

σTd

system

δya

σyb

K

K

acetone + methanol acetone + [emim][triflate] methanol + [emim][triflate] acetone + methanol + [emim][triflate]

0.002

0.002

0.008

0.008

0.01 0.19 0.14 0.13

0.02 0.14 0.19 0.18

a δy ) (1/N) Σ |y b 2 1/2 exptl - ycalcd|. σy ) [1/(N - 1)][Σ(yexptl - ycalcd) ] . δT ) (1/N) Σ |Texptl - Tcalcd|. d σT ) [1/(N - 1)][Σ(Texptl - Tcalcd)2]1/2.

Figure 2. Experimental and calculated y1-x′1 diagrams for acetone (1) + methanol (2) + [emim][triflate] (3) at 100 kPa: dotted line, y1-x′1 calculated IL-free system; dot-dashed line, y1-x′1 calculated for x3 ) 0.055; dashed line, y1-x′1 calculated for x3 ) 0.165; solid line, y1-x′1 calculated for x3 ) 0.360; O, y1-x′1 experimental for x3 ) 0.055; 4, y1-x′1 experimental for x3 ) 0.165; 0, y1-x′1 experimental for x3 ) 0.360.

solvent-IL parameters were determined from the binary vaporliquid equilibrium data of Tables 2 and 3 and the NRTL electrolyte model by minimization of the objective function F F)

∑|T

exptl

- Tcalcd|

(1)

j

Figure 1. Experimental and calculated T-x′1-y1 diagrams for acetone (1) + methanol (2) + [emim][triflate] (3) at 100 kPa: dotted lines, T-x′1-y1 calculated IL-free system; solid lines, T-x′1-y1 calculated for x3 ) 0.055; dot-dot-dashed lines, T-x′1-y1 calculated for x3 ) 0.165; dashed lines, T-x′1-y1 calculated for x3 ) 0.250; dot-dashed lines, T-x′1-y1 calculated for x3 ) 0.360; 9, T-x′1 experimental for x3 ) 0.055; 0, T-y1 experimental for x3 ) 0.055; 2, T-x′1 experimental for x3 ) 0.165; 4, T-y1 experimental for x3 ) 0.165; b, T-x′1 experimental for x3 ) 0.250; O, T-y1 experimental for x3 ) 0.250; 1, T-x′1 experimental for x3 ) 0.360; 3, T-y1 experimental for x3 ) 0.360.

this work, we have used the electrolyte NRTL model to correlate the vapor-liquid equilibrium of the binary system so as to predict the vapor-liquid equilibrium of the ternary system. This model is an extension of the nonrandom two-liquid local composition proposed by Renon and Prausnitz37 from which Chen et al.38 derived a model for single-solvent + electrolyte systems. Subsequent to that, Mock et al.39,40 extended it to mixed-solvent + electrolyte systems, by neglecting the longrange interaction contribution term. In this way, expressions for the liquid-phase activity coefficients of acetone (1) and methanol (2) in a binary or ternary system containing [emim][triflate] (3) can be derived. These equations have been reported elsewhere.41 According to the proposed method, as long as we want to represent the phase equilibrium of mixed-solvent electrolyte systems, we must determine the binary adjustable parameters for all of the solvent-solvent and solvent-electrolyte pairs in the system. Provided that we have three pairs in our system (two solvents and one IL), there are nine binary parameters which have to be calculated. Six of them are energy parameters (∆g1,2, ∆g2,1, ∆g1,3, ∆g3,1, ∆g2,3, and ∆g3,2), and the rest are nonrandomness factors (R1,2 ) R2,1, R1,3 ) R3,1, and R2,3 ) R3,2). The 1-2 binary solvent-solvent parameters were taken from a previous paper.31 On the other hand, the 1-3 and 2-3 binary

where T is the equilibrium temperature, the indices exptl and calcd denote the experimental and calculated values, and the summations are extended to the whole range of data points. All of the nine parameters considered are shown in Table 5, and in Table 6, the mean absolute deviations between the experimental and calculated values of the equilibrium temperature for binary systems can be seen. The parameters reported in Table 5 were used to predict the vapor-liquid equilibrium of the acetone (1) + methanol (2) + [emim][triflate] (3) ternary system, in such a way that we were able to work out the composition in the vapor phase and the equilibrium temperature for each composition in the liquid phase. To do this, ideal behavior of the vapor phase was assumed and the equilibrium conditions represented by eq 2 were considered. yiP ) XiγiP°i

(2)

In eq 2, yi is the vapor-phase mole fraction of solvent i, P is the total pressure in the system, Xi is the liquid-phase mole fraction of component i based on the assumption of total dissociation of the IL, γi is the activity coefficient of component i obtained from the electrolyte NRTL model, and P°i is the vapor pressure of solvent i at equilibrium temperature. The vapor pressures of the pure solvents were calculated by the Antoine equation using the parameters obtained with the same recirculating still.31 The ability of the model to reproduce the vapor-liquid equilibrium for this system can be seen in Figure 1 where the calculated and experimental vapor-liquid equilibrium of the acetone + methanol + [emim][triflate] ternary system is plotted on a T-x′1-y1 diagram for x3 ) 0, 0.055, 0.165, 0.250, and 0.360. This figure and the mean absolute deviations from Table 6 reveal that the proposed model successfully fits the experimental data, and only for the highest [emim][triflate] concentrations can small differences be appreciated, despite the fact that

Journal of Chemical and Engineering Data, Vol. 52, No. 1, 2007 145 Table 7. Minimum Value of Mole Fraction of Entrainer x3 Which Is Needed to Break the Acetone + Methanol Azeotrope entrainer

x3

equilibrium conditions

ref

sodium iodide calcium bromide lithium nitrate sodium thiocyanate potassium thiocyanate zinc chloride [emim][triflate] lithium bromide [emim][BF4]

0.015a

isobaric, P ) 101.32 kPa isobaric, atmospheric pressure isobaric, P ) 100.00 kPa isobaric, P ) 101.32 kPa isobaric, P ) 101.32 kPa isobaric, P ) 101.32 kPa isobaric, P ) 100.00 kPa isotherm, T ) 343.75 K isotherm, T ) 328.15 K

Iliuta and Thyrion44 Al-Asheh and Banat48 Vercher et al.31 Iliuta and Thyrion45 Iliuta et al.46 Yan et al.49 this work Yan et al.47 Seiler et al.8

0.021a 0.022b 0.030a 0.050a 0.050a 0.078b 0.050a 0.100a

a Experimental value, selected from a discrete set, in which it is observed that the azeotrope is broken. b Estimated value by modeling, from which the azeotrope is broken.

Figure 3. Salting-out effect of [emim][triflate] and lithium nitrate on the vapor-liquid equilibrium of the acetone (1) + methanol (2) system at 100 kPa for an electrolyte mole fraction of x3 ) 0.120: solid line, [emim][triflate]; dashed line, lithium nitrate; dotted line, salt-free system.

an [emim][triflate] mole fraction of 0.36 represents a mass fraction as high as 72-82%. It is worth noting that small concentrations of [emim][triflate] produce a displacement of the azeotropic point of the acetone + methanol system toward x′1 values higher than 0.781 until the azeotrope disappears, as seen in Figure 2. As a matter of fact, it is seen in Table 4 that an x3 mole fraction of 0.055 in IL is very closed to the limit from which the azeotrope disappears. From the NRTL electrolyte model, it can be estimated that the disappearance of the azeotropic point for the acetone + methanol system at 100 kPa takes place at x3 ) 0.078. This IL concentration is higher than that obtained when an inorganic salt was used, as can be seen in Table 7 where the minimum mole fraction of the electrolyte to break the acetone + methanol azeotrope is reported. As far as we know, only the action of an ionic liquid (1-ethyl-3-methylimidazolium tetrafluoroborate)8 on the acetone + methanol system has been reported, but at isotherm conditions (T ) 328.15 K). In this case, the azeotrope disappears at x3 ) 0.100. It is a generally accepted fact that the variations of salt solubility in a mixed solvent with the system composition indicate that the salt interacts more strongly with one of solvents; that is, the salt effect is more pronounced, and thus using this salt as an entrainer in an extractive distillation, we can improve solvent separation. In accordance to this, [emim][triflate], which is totally miscible in acetone and methanol, should present a relative weak salt effect. However, what this great solubility in the mixture does is ensure an IL mole fraction in the liquid phase as high as one wants, and therefore, the salt effect, which is proportional to the amount of electrolyte that is present in solution, can be augmented. So, these unique properties of ILs

combining the ionic properties of inorganic salts with its large solubility even in nonpolar solvents make them very suitable as entrainers in extractive distillation. For instance, the addition of lithium chloride in the acetone + methanol mixture at atmospheric pressure results in a very large increase in the acetone mole fraction of the vapor phase. However, the azeotrope of the mixed solvent system does not disappear even at saturation because the salt solubility in the mixture decreases dramatically as the acetone composition increases.42 In this way, it is the decreased solubility of the inorganic salts in low-polar solvents that appears as its greatest handicap. Conversely, ILs do not present this sort of problem. In the acetone + methanol + [emim][triflate] system, a crossover effect43 between salting-in and salting-out of ILs on acetone is observed, such as in the case of sodium iodide, potassium iodide,44 sodium thiocyanate,45 or potassium thiocyanate.46 On the contrary, this effect does not arise when lithium chloride,42 lithium bromide,47 lithium nitrate,31 calcium bromide,48 or zinc chloride49 is used. Figure 3 shows the effect of [emim][triflate] and lithium nitrate31 on the vapor-liquid equilibrium of the acetone + methanol system for an electrolyte mole fraction of x3 ) 0.120. The different behavior of the two electrolytes can be observed in it.

Conclusions In this work, it has been confirmed that the electrolyte NRTL model, which has been extensively used with inorganic salts, is also suitable to predict the vapor-liquid equilibrium in the presence of an organic salt such as [emim][triflate]. In this way, this will allow us to extend the application of the model to the promising field of ionic liquids. The addition of IL [emim][triflate] to acetone + methanol originates a noticeable salting-out effect on acetone near the azeotropic point and a slight salting-in effect at lower acetone concentrations (crossover effect), although this does not influence in any way the complete separation of these solvents. From the vapor-liquid equilibrium viewpoint, [emim][triflate] is as effective as other inorganic salts in breaking the azeotrope of the acetone + methanol system. If we think about the extractive distillation process of acetone + methanol mixtures, the use of [emim][triflate] as the entrainer presents obvious advantages over inorganic salts. A great concentration of the electrolyte can be achieved in the liquid phase along the column because of its total miscibility with both solvents. Moreover, [emim][triflate] can be totally separated from methanol by flash distillation of the column bottom’s stream, and a pure [emim][triflate] stream can be easily handled and mixed with the reflux stream to go back to the column top without introducing any additional amounts of methanol. Compared with classic liquid entrainers, ionic liquids have two advantages: their ionic character, which produces heavier

146

Journal of Chemical and Engineering Data, Vol. 52, No. 1, 2007

interactions with solvents and therefore a greater separating effect, and their zero vapor pressure which allows us to recover and reuse them free of a less-volatile component. Furthermore, although wastage by evaporation is null, a considerable amount of ionic liquid is needed inside the column to break the azeotrope because of its large molecular mass. This fact and the high price of ionic liquids constitute the highest disadvantages. Nevertheless, we are confident the ionic liquids will come down in price when its demand augments, and what is more, one should not forget that the ionic liquid consumption in the distillation process is zero because its recovery is complete.

Literature Cited (1) Brennecke, J. F.; Maginn, E. J. Ionic Liquids: Innovative Fluids for Chemical Processing. AIChE. J. 2001, 47, 2384-2389. (2) Seddon, K. R. Ionic Liquids For Clean Technology. J. Chem. Technol. Biotechnol. 1997, 68, 351-356. (3) Heintz, A. Recent developments in thermodynamics and thermophysics of non-aqueous mixtures containing ionic liquids. A review. J. Chem. Thermodyn. 2005, 37, 525-535. (4) Swatloski, R. P.; Visser, A. E.; Reichert, W. M.; Broker, G. A.; Farina, L. M.; Holbrey, J. D.; Rogers, R. D. On the solubilization of water with ethanol in hydrophobic hexafluorophosphate ionic liquids. Green Chem. 2002, 4, 81-87. (5) Arlt, W.; Seiler, M.; Jork, C.; Schneider, T. DE Patent No. 10114734, 2001. (6) Arlt, W.; Seiler, M.; Jork, C.; Schneider, T. DE Patent No. 10136614, 2001. (7) Arlt, W.; Seiler, M.; Jork, C.; Schneider, T. Ionic Liquids as Selective Additives for the Separation of Close-Boiling or Azeotropic Mixtures. PCT Int. Appl., 2002, WO 02/074718 A2. (8) Seiler, M.; Jork, C.; Schneider, W.; Arlt, W. Ionic liquids and hyperbranched polymers - Promising new classes of selectiVe entrainers for extractiVe distillation. In Proceedings of the International Conference on Distillation & Absorption, 2002; GVC-VDI: Du¨sseldorf, 2002. (9) Arlt, W.; Seiler, M.; Jork, C. New classes of compounds for Chemical and Process Engineering: ionic liquids and hyperbranched polymers. In 6th Italian Conference on Chemical and Process Engineering; Pierucci, S., Ed.; Chem. Eng. Transactions 2003, 3, 1257-1269. (10) Seiler, M.; Jork, C.; Kavarnou, A.; Arlt, W.; Hirsch, R. Separation of Azeotropic Mixtures Using Hyperbranched Polymers or Ionic Liquids. AIChE J. 2004, 50, 2439-2454. (11) Jork, C.; Seiler, M.; Beste, Y. A.; Arlt, W. Influence of Ionic Liquids on the Phase Behavior of Aqueous Azeotropic Systems. J. Chem. Eng. Data 2004, 49, 852-857. (12) Beste, Y.; Eggersmann, M.; Schoenmakers, H. Extractive distillation with ionic fluids. Chem. Ing. Tech. 2005, 77, 1800-1808. (13) Lei, Z.; Arlt, W.; Wasserscheid, P. Separation of 1-hexene and n-hexane with ionic liquids. Fluid Phase Equilib. 2006, 241, 290299. (14) Anthony, J. L.; Maginn, E. J.; Brennecke, J. F. Solution Thermodynamics of Imidazolium-Based Ionic Liquids and Water. J. Phys. Chem. B 2001, 105, 10942-10949. (15) Verevkin, S. P.; Vasiltsova, T. V.; Bich, E.; Heintz, A. Thermodynamic properties of mixtures containing ionic liquids. Activity coefficients of aldehides and ketones in 1-methyl-3-ethyl-imidazolium bis(trifluoromethylsulfonyl) imide using the transpiration method. Fluid Phase Equilib. 2004, 218, 165-175. (16) Verevkin, S. P.; Safarov, J.; Bich, E.; Hassel, E.; Heintz, A. Thermodynamic properties of mixtures containing ionic liquids. Fluid Phase Equilib. 2005, 236, 222-228. (17) Vasiltsova, T. V.; Verevkin, S. P.; Bich, E.; Heintz, A.; Bogel-Lukasik, R.; Domaska, U. Thermodynamic Properties of Mixtures Containing Ionic Liquids. 7. Activity Coefficients of Aliphatic and Aromatic Esters and Benzylamine in 1-Methyl-3-ethylimidazolium Bis(trifluoromethylsulfonyl) Imide Using the Transpiration Method. J. Chem. Eng. Data 2006, 51, 213-218. (18) Heintz, A.; Verevkin, S. P.; Ondo, D. Thermodynamic Properties of Mixtures Containing Ionic Liquids. 8. Activity Coefficients at Infinite Dilution of Hydrocarbons, Alcohols, Esters, and Aldehydes in 1-Hexyl3-methylimidazolium Bis(trifluoromethylsulfonyl) Imide Using GasLiquid Chromatography. J. Chem. Eng. Data 2006, 51, 434-437. (19) Safarov, J.; Verevkin, S. P.; Bich, E.; Heintz, A. Vapor Pressures and Activity Coefficients of n-Alcohols and Benzene in Binary Mixtures with 1-Methyl-3-butylimidazolium Octyl Sulfate and 1-Methyl-3octylimidazolium Tetrafluoroborate. J. Chem. Eng. Data 2006, 51, 518-525.

(20) Heintz, A.; Vasiltsova, T. V.; Safarov, J.; Bich, E.; Verevkin, S. P. Thermodynamic Properties of Mixtures Containing Ionic Liquids. 9. Activity Coefficients at Infinite Dilution of Hydrocarbons, Alcohols, Esters, and Aldehydes in Trimethyl-butylammonium Bis(trifluoromethylsulfonyl) Imide Using Gas-Liquid Chromatography and Static Method. J. Chem. Eng. Data 2006, 51, 648-655. (21) Kim, K. S.; Shin, B. K.; Lee, H.; Ziegler, F. Refractive index and heat capacity of 1-butyl-3-methylimidazolium bromide and 1-butyl3-methylimidazolium tetrafluoroborate, and vapor pressure of binary systems for 1-butyl-3-methylimidazolium bromide + trifluoroethanol and 1-butyl-3-methylimidazolium tetrafluoroborate + trifluoroethanol. Fluid Phase Equilib. 2004, 218, 215-220. (22) Kim, K. S.; Park, S. Y.; Choi, S.; Lee, H. Vapor Pressures of the 1-Butyl-3-methylimidazolium Bromide + Water, 1-Butyl-3-methylimidazolium Tetrafluoroborate + Water, and 1-(2-Hydroxyethyl)3-methylimidazolium Tetrafluoroborate + Water Systems. J. Chem. Eng. Data 2004, 49, 1550-1553. (23) Kato, R.; Krummen, M.; Gmehling, J. Measurement and correlation of vapor-liquid equilibria and excess enthalpies of binary systems containing ionic liquids and hydrocarbons. Fluid Phase Equilib. 2004, 224, 47-54. (24) Do¨ker, M.; Gmehling, J. Measurement and prediction of vapor-liquid equilibria of ternary systems containing ionic liquids. Fluid Phase Equilib. 2005, 227, 255-266. (25) Kato, R.; Gmehling, J. Measurement and correlation of vapor-liquid equilibria of binary systems containing the ionic liquids [EMIM][(CF3SO2)2N], [BMIM][(CF3SO2)2N], [MMIM][(CH3)2PO4] and oxygenated organic compounds respectively in water. Fluid Phase Equilib. 2005, 231, 38-43. (26) Kato, R.; Gmehling, J. Systems with ionic liquids: Measurement of VLE and γ∞ data and prediction of their thermodynamic behavior using original UNIFAC, mod. UNIFAC(Do) and COSMO-RS(Ol). J. Chem. Thermodyn. 2005, 37, 603-619. (27) Abusen, Z. R.; Zhao, J.; Li, C.; Wang, Z. Determination of vapor pressures for binary and ternary mixtures containing ionic liquid 1-propyl-3-methylimidazolium bromide. Chin. J. Chem. Eng. 2005, 13, 791-795. (28) Orchille´s, A. V.; Gonza´lez-Alfaro, V.; Vercher, E.; Martı´nez-Andreu, A. Vapor pressures of ionic liquid 1-butyl-3-methylimidazolium octylsulfate [BMIM] [OcOSO3] in water and 1-propanol. Proceedings of the 10th Mediterranean Congress of Chemical Engineering; SEQIIQ: Barcelona, 2005. (29) Banerjee, T.; Singh, M. K.; Khanna, A. Prediction of Binary VLE for Imidazolium Based Ionic Liquid Systems Using COSMO-RS. Ind. Eng. Chem. Res. 2006, 45, 3207-3219. (30) Zhao, J.; Dong, C. C.; Li, C. X.; Meng, H.; Wang, Z. H. Isobaric vapor-liquid equilibria for ethanol-water system containing different ionic liquids at atmospheric pressure. Fluid Phase Equilib. 2006, 242, 147-153. (31) Vercher, E.; Orchille´s, A. V.; Miguel, P. J.; Gonza´lez-Alfaro, V.; Martı´nez-Andreu, A. Isobaric vapor-liquid equilibria for acetone + methanol + lithium nitrate at 100 kPa. Fluid Phase Equilib. 2006, doi:10.1016/j.fluid.2006.09.007. (32) Walas, S. M. Phase Equilibria in Chemical Engineering; Butterworth: London, 1985. (33) Harper, B. G.; Moore, J. C. Vapor-Liquid Equilibrium. New Still and Method for Determining Vapor-Liquid Equilibrium. Ind. Eng. Chem. 1957, 49, 411-414. (34) Freshwater, D. C.; Pike, K. A. Vapor-Liquid Equilibrium Data for Systems of Acetone-Methanol-Isopropanol. J. Chem. Eng. Data 1967, 12, 179-183. (35) Van Ness, H. C.; Byer, S. M.; Gibbs, R. E. Vapor-Liquid Equilibrium: Part I. An Appraisal of Data Reduction Methods. AIChE J. 1973, 19, 238-244. (36) Fredenslund, A.; Gmehling, J.; Rasmussen, P. Vapor-Liquid Equilibria Using UNIFAC; Elsevier: Amsterdam, 1977. (37) Renon, H.; Prausnitz, J. M. Local Compositions in Thermodynamic Excess Functions for Liquid Mixtures. AIChE J. 1968, 14, 135-144. (38) Chen, C. C.; Britt, H. I.; Boston, J. F.; Evans, L. B. Local Composition Model for Excess Gibbs Energy of Electrolyte Systems. Part I: Single Solvent, Single Completely Dissociated Electrolyte Systems. AIChE J. 1982, 28, 588-596. (39) Mock, B.; Evans, L. B.; Chen, C. C. Phase equilibria in multiplesolvent electrolyte systems: a new thermodynamic model. Proc. Summer Comput. Simul. Conf. 1984, 558-562. (40) Mock, B.; Evans, L. B.; Chen, C. C. Thermodynamic Representation of Phase Equilibria of Mixed-Solvent Electrolyte Systems. AIChE J. 1986, 32, 1655-1664. (41) Vercher, E.; Rojo, F. J.; Martı´nez-Andreu, A. Isobaric Vapor-Liquid Equilibria for 1-Propanol + Water + Calcium Nitrate. J. Chem. Eng. Data 1999, 44, 1216-1221.

Journal of Chemical and Engineering Data, Vol. 52, No. 1, 2007 147 (42) Iliuta, M. C.; Iliuta, I.; Landauer, O. M.; Thyrion, F. C. Salt effect of LiCl on vapor-liquid equilibrium of the acetone-methanol system. Fluid Phase Equilib. 1998, 149, 163-176. (43) Meranda, D.; Furter, W. F. Salt Effects on VLE: Some Anomalies. AIChE J. 1974, 20, 103-108. (44) Iliuta, M. C.; Thyrion, F. C. Vapour-liquid equilibrium for the acetone-methanol-inorganic salt system. Fluid Phase Equilib. 1995, 103, 257-284. (45) Iliuta, M. C.; Thyrion, F. C. Salt effects on vapour-liquid equilibrium of acetone-methanol system. Fluid Phase Equilib. 1996, 121, 235252. (46) Iliuta, M. C.; Thyrion, F. C.; Landauer, O. M. Vapour-liquid equilibrium of the acetone-methanol system in the presence of KSCN and comparison with other salts. Fluid Phase Equilib. 1997, 130, 253269. (47) Yan, W.; Rose, C.; Zhu, M.; Gmehling, J. Measurement and Correlation of Isothermal Vapor-Liquid Equilibrium Data for the

System Acetone + Methanol + Lithium Bromide. J. Chem. Eng. Data 1998, 43, 585-589. (48) Al-Asheh, S.; Banat, F. Isobaric Vapor-Liquid Equilibrium of Acetone + Methanol System in the Presence of Calcium Bromide. J. Chem. Eng. Data 2005, 50, 1789-1793. (49) Yan, W.; Topphoff, M.; Zhu, M.; Gmehling, J. Measurement and Correlation of Isobaric Vapor-Liquid Equilibrium Data for the System Acetone + Methanol + Zinc Chloride. J. Chem. Eng. Data 1999, 44, 314-318.

Received for review July 14, 2006. Accepted September 27, 2006. Financial support by the Spain Ministry of Education and Culture (Grant CTQ2004-02977/PPQ) is gratefully acknowledged.

JE0603170